CPU Temps; I am out of Ideas!

macklin01 said:

Destroyer,

Thanks for your response. I'm not sure. The temp changes have always been fairly instantaneous with my CPU readings in MBM5, which makes me lean towards internal diode. It may well depend on teh version of PIII, too. (I have FCPGA-2, not a slot version.) Also, I can see no trace of a thermistor on the motherboard.

Click to expand...

Mine was a Flip chip to on an Abit BF-6. Actually I've noticed near instantaneous responses to temperature regardless of whether its on chip or using thermal diode as long as the refresh interval of MBM is set to minimum.(1 sec?)

Out of curiosity, I emailed the motherboard maker to ask. I'll let you know if I hear back from them. Always interesting stuff to find out ....

Click to expand...

This would be good! I hunger for the knowledge.

Oh, and I don't think he has a better HSF; His temps are on the order of 10C higher, which is consistent with the higher heat buildup you sensed on his AMD system as opposed to yours.

Click to expand...

Actually I was comparing his 40-50+C temps with my 50-60C temps on my P4. He has an average cheapo coolermaster and I have my TT Vol7+. All things being equal, my CPU ought to radiate more heat than his due to the fact that mine has better heat dissipation.

The fact that his feels hotter and yet has a lower temp reading suggests that his temp sensor is probably in a less sensitive location.

In a way I feel that temperature comparisons across different makes/brands of mobo are not comparable unless we are sure where the readings are taken from.

I've assembled a few AXP systems and I cannot reconcile the MBM temps recorded with the heat felt by the trusty back of my hand.

Incidentally, can anyone confirm if Asus P4S333 owners get lower temps? I did one up for my friend recently and he was getting 10deg lower temps at 2.4Ghz/stock HSF+ cheap casing than mine at 2.3Ghz/Vol7+ badass casing.

I get a little ****ed.

Mike360000 said:

Still the idle temps has me stymied.... Why doesn't my idle temps go down? (I ran a 24 loop Sandra cpu burn in to check the system.)

Click to expand...

I'd think this is a function of the small die size of the P4. Though it has the IHS, it can barely help only if well applied.

A larger thermal gradient would result in higher rate of heat transfer, that is why the difference in temps are greater only for full load measurements.

To reduce your idle temps, you will need to either drastically improve thermal conductivity of the material or coupling(implausible) or perhaps lower the ambient temperature of to get a higher thermal gradient.

Zerstorer,
"Geez Mike,

You case temp is way too high for a setup with such good ventilation. Have you tried increasing the exhaust flow? I was thinking that maybe you have too much positive air pressure?

Coz my case temps are usually only 3degrees above ambient temp. I live in the tropics with 30deg C room temperatures, case temp is usually 33-34C."

Yep I tried with a side cover off and with side cover on and one of the 5 1/4" drive bay covers removed. I couldn't really tell any difference either way! Maybe 1 or 2c at best. One would have at least thought the best temps, under load, would have been had with the side cover off! I've always believed and have just recently read at other review sites, that it is a little better to have a little more positive than negative air pressure. My 8" fan isn't a real powerful one, but it does pressurize the system. I am really beginning to think that most of the BD7 mobos have a tendency to run hotter or show hotter temps tham most other mobos. I also believe at this point, my best bet is too try baffling the air flow, directing it towards the motherboard, and then maybe increasing the airflow with the bigger fan I mentioned earlier.

I just played UT for about 45 minutes, but quit when my system alarm went off. It was timed at 48c. CPU read good though, at 58c! At least that is one noticeable improvement......

I'm also thinking about upgrading my BIOS, as I have an earlier version.

Cheers,
Mike

Zerstorer said:

Incidentally, can anyone confirm if Asus P4S333 owners get lower temps? I did one up for my friend recently and he was getting 10deg lower temps at 2.4Ghz/stock HSF+ cheap casing than mine at 2.3Ghz/Vol7+ badass casing.

Click to expand...

Hi,

just wondered what temps you all are getting with the Intel stock cooling. I use some ShinEtsu thermal grease (similar to Coolermaster Premium) and I have idle temps at 39-40 degrees C. and full load 44.

My mainboard is idle at 32 and full load 37. Temps do not change when I overclock to 105MHz FSB (1.9GHz). I have two case fans which brought down temps 3-5.

I have Windows XP Pro, so the HLT-command will cool my CPU down when not in use.

But I see that some do have idle temps of 31 for their CPUs even at a higher clock!

In this thread http://forums.anandtech.com/message...188&STARTPAGE=1 there are people reporting this:

"Mine runs 30-32 idle.. at 2.5Ghz and default voltage (from 1.6A Northwood using Intel retail Heatsink)... after 15mins of full load it will jump to 38c temp reading comes from Asus PC Probe 2.16.04 (I think)"

How is that possible?? There are all using Asus boards with PC Probe. It is that why they get that low temps??

Merkor

The last time I used the Intel HSF I remembered getting temps of 43-48. With 1 deg higher similar case temps as yours, at default vcore.

Btw the link you posted is dead.

Zerstorer said:

This would be good! I hunger for the knowledge.

Click to expand...

Still no response. I'll try again later. -- Paul

Response on PIII monitoring from Shuttle

Date: Tue, 11 Jun 2002 11:36:14 -0700
From: Shuttle Tech Dept. <[email protected]>
To: Paul Macklin <[email protected]>
Subject: Re: question about AE25 temp monitoring

The processor temperature is read off onboard sensor, not from your CPU
internal Diode.

As you as you can your processor temperature around 45 Celsius - 55
Celsius, your are Okay.

( Is calibration necessary?) You need to explain more in detail regards calibration for your processors temp. Or call us for further assistances.

xxxx
Shuttle Computer Group
Technical Support Dept.
939 Radecki Court
Industry CA 91748
(626)820-9000 ex.xxxx, xxxx, xxxx

Re: Response on PIII monitoring from Shuttle

macklin01 said:

Date: Tue, 11 Jun 2002 11:36:14 -0700
From: Shuttle Tech Dept. <[email protected]>
To: Paul Macklin <[email protected]>
Subject: Re: question about AE25 temp monitoring

The processor temperature is read off onboard sensor, not from your CPU
internal Diode.

As you as you can your processor temperature around 45 Celsius - 55
Celsius, your are Okay.

( Is calibration necessary?) You need to explain more in detail regards calibration for your processors temp. Or call us for further assistances.

xxxx
Shuttle Computer Group

Technical Support Dept.
939 Radecki Court
Industry CA 91748
(626)820-9000 ex.xxxx, xxxx, xxxx

Click to expand...

The reference That I made to calibrating your temperatures was specificaly for motherboards that use the internal diode as per the link to Artic Silver.com. I do not know if the calibration would be the same for your on board sensor temps. Usualy the on board temps are 7c to 10c lower and this is what most people refer to when posting there load temps. You might send an email to them and see what they say.

Re: Re: Response on PIII monitoring from Shuttle

davefred99 said:

The reference That I made to calibrating your temperatures was specificaly for motherboards that use the internal diode as per the link to Artic Silver.com. I do not know if the calibration would be the same for your on board sensor temps. Usualy the on board temps are 7c to 10c lower and this is what most people refer to when posting there load temps. You might send an email to them and see what they say.

Click to expand...

Thanks. Actually, once he emailed to say that it was an onboard sensor, I figured that the calibration was pretty much moot. I did email, though, with a link to the article to hear what he had to say.

Thanks! -- Paul

bump

Here are my pics of my baffle and squirrel cage fan.
Cheers,
Mike

And my squirrel cage fan. --Mike

bump

At least I'm not the only one experiencing this problem I guess. I'm also running a P4 2.26G "B" on an Abit BD7-II. My ambient temp. right by my comp. (as measured with a Thermocouple attached to a Fluke temp. meter) is 30.6 C. MBM5 is measuring my case temp. at 41 C and my cpu at 45 C. I top out (cpu) at around 53/54 C at full load. I've got 2-92mm fans and 1-80mm fan blowing in and 1-92mm and 2-80mm exhaust fans. This is in an Antec sx1030b case. I'm using a Swiftech 4000 on my cpu. Both the heatsink and the heatspreader have been lapped. Using AS3. Even considering my room temp. of 30/31 C I should be cooler at my case and my cpu. I think the temps are measured too high. I also have e-mailed Abit regarding this problem. No response as of yet. If anyone gets any resolution on this issue please post immediately. Thanks.

Try reading your case temps with the case door or cabinet off.
If it varies by several degrees, your mobo is probably
reading ok. The mobo temp is read at the MTH chip, and it
stays pretty warm around that spot, with airflow a little restricted. I wouldn't hold my breath waiting for a
reply from Abit. I've never gotten anything back from them,
althoug others have.

Cheers,
Mike

ABIT's temps are screwed. Same thing happens on certain TH7IIs, and on all BD7, BD7II, IT7, IT7-MAX2.